Merge pull request #3002 from yunkunrao/add_kubelet_pleg_relist_quantile

Add node metric: node_pleg_quantile

pkg/models/monitoring/named_metrics.go

@@ -112,6 +112,7 @@ var NodeMetrics = []string{
 	"node_load5",
 	"node_load15",
 	"node_pod_abnormal_ratio",
+	"node_pleg_quantile",
 }
 
 var WorkspaceMetrics = []string{

pkg/simple/client/monitoring/prometheus/promql.go

@@ -109,6 +109,7 @@ var promQLTemplates = map[string]string{
 	"node_load5":                  `node:load5:ratio{$1}`,
 	"node_load15":                 `node:load15:ratio{$1}`,
 	"node_pod_abnormal_ratio":     `node:pod_abnormal:ratio{$1}`,
+	"node_pleg_quantile":          `node_quantile:kubelet_pleg_relist_duration_seconds:histogram_quantile{$1}`,
 
 	// workspace
 	"workspace_cpu_usage":                  `round(sum by (workspace) (namespace:container_cpu_usage_seconds_total:sum_rate{namespace!="", $1}), 0.001)`,